1. Field of the Invention
The present invention relates to a transmission power control apparatus which controls transmission power to keep a communication quality of user information at a predetermined quality.
2. Description of the Related Art
Conventionally, a mobile radio communication system in which radio communications are performed between a base station and a mobile station is developed and put to practical use. Particularly, a CDMA communication system has advantages that the spectral efficiency and the system capacity are high because a plurality of communication links are available in the same frequency band by identifying each communication link with a respective spreading code.
As illustrated in FIG. 1, base station BS communicates data with mobile stations MS1 and MS2 using respective reverse links and forward links. Information bits transmitted in each communication link are spectrum spread with different spreading codes.
Received levels of communication link signals in the CDMA communication system vary according to time due to a distance between the base station and the mobile station and fading. The interference noise level varies similarly. FIG. 2 illustrates an example of a variation of Signal to Interference Ratio (SIR). In the CDMA radio communication system, the greatest system capacity is obtained when each SIR of both communication link signals is equal to each other. Therefore, base station BS performs transmission power control in order to keep a received level of each communication link signal at the same level.
An example is explained using the reverse link. Base station BS receives a reverse link signal and measures a received SIR. Base station BS further compares the measured result with a reference SIR, and instructs mobile station MS to increase transmission power when the measured result is lower than the reference value, while instructs to decrease transmission power when the measured result is higher than the reference value. The instruction is transmitted to mobile station MS via forward link. The received SIR in base station BS is thus controlled at a value around the reference SIR (High rate transmission power control).
FIG. 3 illustrates an example of a relation between received SIR and communication link Bit Error Rate (hereinafter referred to as BER). The characteristics vary depending on propagation environments such as moving speed. In order to obtain communication link BER of 0.1%, environment A requires reference value A as a reference SIR, and environment B requires reference value B as a reference SIR. Since it is difficult to estimate propagation environments, actually performed is that error detection bits such as CRC are added to user information, and a reception side detects presence of absence of error and measures BER or Frame Error Rate (hereinafter referred to as FER) per a frame. When the measured BER or FER is lower than a predetermined value for a desired communication quality, reference SIR is decreased, and when the measured BER or FER is lower than the predetermined value, reference SIR is increased, in order to keep a communication quality of user information at a predetermined quality (Low rate transmission power control).
On the other hand, a concatenated code, in which a plurality of error correcting codes are combined, is used in high quality data transmissions. FIG. 4 illustrates an exemplary diagram for coding and decoding of the concatenated code obtained by combining convolutional code and Read Solomon (RS) code. A transmission side performs RS coding on user information, further performs convolutional coding on the RS coded data, and then modulates the convolutional coded data to transmit by radio. A reception side performs Viterbi decoding of the convolutional coded data, which is obtained by demodulating received signals, and performs RS decoding of the decode data to obtain the user information.
The reception side performs Viterbi decoding of received convolutional coded data for error correction to obtain data with BER of approximately 10E-4, and further RS decoding of the decoded data for error correction to obtain data with BER of approximately 10E-6.
A following example illustrates the case where user information is transmitted at a data rate of 64 kbps with a quality of approximately BER=10E-6. Since it is difficult to measure BER, FER is measured using error detection code such as CRC. When 1 frame has 100 bits, the number of frames per second is 720 [frames/sec]. Approximately, BER=10E-4 corresponds to FER=10E-3 with error burst characteristics considered. In this case, since 7,200 frames are transmitted for 10 seconds, the number of error frames is approximately 7 for 10 seconds, making it possible to measure FER with accuracy of one digit. When long term transmission power control is performed based on this FER value, a time constant is approximately 10 seconds.
FIG. 5 illustrates a configuration for receiving a concatenated code obtained by combining convolutional code and RS code and then generating a command for transmission power control. SIR measurement section 1 measures SIR of the received signal, while Viterbi decoding section 2 performs Viterbi decoding of the received signal. RS decoding section 4 performs RS decoding of Viterbi decoded CC coded data to output user information.
Further, FER measurement section 3 measures FER of the CC coded data output from Viterbi decoding section 2. Reference SIR control section 5 inputs reference SIR to comparison section 6, while controlling the reference SIR corresponding to FER. Comparison section 6 determines whether transmission power is increased or decreased comparing the reference SIR to measured SIR so as to generate a TPC command.
Since the concatenated code as described above has a 2-stage structure, it is possible to perform the FER measurement at the time Viterbi decoding is finished prior to RS decoding, enabling the time constant to be shortened.
Further, a turbo code is recently paid attention as a code with high error correction capability. The turbo code is summarized in xe2x80x9cIterative Decoding of Binary Block and Convolutional Codesxe2x80x9d by J. Hagenauer et al. (IEEE TRANSACTION ON INFORMATION THEORY, Vol.42, No.2, March 1996).
FIG. 6 illustrates a schematic configuration of a turbo decoder, to which a received signal is input and from which a decoding result of soft value is output. The turbo decoder calculates communication path value (1), previous likelihood (2), and external information likelihood (3), and outputs a sum of (1), (2) and (3). In the first decoding, the previous likelihood is set at 0. From the second decoding, the previous likelihood is updated to previous external information likelihood and the same calculation is performed. Such processing is iterated corresponding to the predetermined number of iterations, while the output is updated. FIG. 7 illustrates an example of BER when the iterative decoding is performed. As can be seen from FIG. 7, BER is decreased as the decoding is performed iteratively even when SIR is the same.
FIG. 8 illustrates a configuration for receiving the turbo code and then generating a command for transmission power control. SIR measurement section 11 measures SIR of the received signal, while turbo decoding section 12 performs iterative decoding of the received signal. The decoding result obtained after a predetermined number of iterations, which times a desired FER is obtained, is output as user information.
Further, FER measurement section 13 receives the decoding result output from turbo decoding section 12 as the user information to measure FER. Reference SIR control section 14 inputs the reference SIR to comparison section 15, while controlling the reference SIR corresponding to FER. Comparison section 15 determines whether transmission power is increased or decreased comparing the reference SIR to a measured SIR so as to generate a TPC command.
FIG. 9 illustrates an exemplary diagram for coding and decoding of the turbo code. In the case of using one kind of error correcting code such as a turbo code, a bit rate is decreased from 192 kbps to 64 kbps by one step according to the turbo decoding, not by multiple steps stepwisely, at a reception side. FER is measured from the signal with a bit rate of 64 kbps.
However, in the case of using one kind of error correcting code such as the turbo code not a concatenated code, a time constant for long term (low rate) transmission power control becomes long, thereby remains the problem that it is difficult to keep stable communication quality of user information.
A specific example is explained in the case of measuring a quality of user information data. Assume that FER is 10E-4 when BER is approximately 10E-6, and that 1 frame has 300 bits. The number of frames per 10 seconds is 2133, and the number of error frames per 10 seconds is 0.2 as an average value. However, since the number of error frames takes an integer number, the average value for 10 seconds is replaced with 0 in this case, or 1, 2 or etc., resulting in a great error between an actual average value and a quantized value. Therefore, it is difficult to measure FER with accuracy of one digit, and several hundred seconds are necessary to measure FER with accuracy of one digit. In other words, several hundred seconds are necessary as a time constant to perform the long term transmission power control based on FER in the case of using the turbo code. When a time constant is long, it may take a long time to detect a variation after a communication path propagation environment varies. It is thereby difficult to keep stable communication qualities.
An object of the present invention is to provide a transmission power control apparatus capable of keeping communication qualities of user information stably without extending a time constant for long term transmission power control even when a single kind of error correcting code such as a turbo code is used.
The present invention provides a transmission power control apparatus which detects a data quality after a smaller number of decoding iteration times, and based on the detected quality, performs transmission power control when a high quality data transmission is performed using an error correcting code with the iterative decoding performed.
According to the aforementioned apparatus, it is possible to keep communication qualities of user information stably without extending a time constant for long term transmission power control even when a single kind of error correcting code such as the turbo code is used.